The Coase Theorem, Transaction Costs and Forward Induction
-
26 Jan 2027
12:45-14:00, Butler Room, Nuffield College
- Economic Theory Lunchtime Workshop Add to Calendar
University of Oxford
Abstract: The Coase Theorem is widely interpreted to imply that, in the absence of private information, bargaining between just two parties yields a Pareto-efficient outcome. This paper asks whether that conclusion follows from rationality alone when, as with purely private transactions, there is no pre-specified bargaining protocol the parties must use. Using forward induction and Mertens-stable sets, a refinement of Nash equilibrium, we characterize outcomes when parties can take unilateral actions that reduce their own payoffs (“burn utility” in the game theory terminology). If only one party can burn utility, the unique Mertens-stable outcome involves no burning of utility and is Pareto efficient. If both parties can do so, every stable outcome involves burning utility with positive probability, generating an endogenous Williamson-type transaction cost despite complete information. This inefficiency might be eliminated in contexts where a shared social bargaining norm has developed or both parties are part of an institution, such as a firm, that imposes a protocol with a unique efficient equilibrium. Thus, the applicability of the Coase Theorem depends on empirically observable features of the environment.
The Economic Theory Lunchtime Workshops are convened by Meg Meyer.